Design and Performance Evaluation of Configurable Component Middleware for End-to-End Adaptation of Distributed Real-Time Embedded Systems

Shankaran, N.; Schmidt, Douglas C.; Koutsoukos, Xenofon; Chen, Yingming; Lu, Chenyang

doi:10.1109/isorc.2007.27

Cited by 11 publications

(9 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition to real-time Java VMs, there is a trend towards the use of middleware to support real-time programming [25,31]. Such real-time middleware systems include real-time CORBA implementations [28,27] and real-time CLI environments [12].…”

Section: Jtres'09mentioning

confidence: 99%

Avoiding unbounded priority inversion in barrier protocols using gang priority management

Röck

Auerbach²,

Kirsch

et al. 2009

Proceedings of the 7th International Workshop on Java Technologies for Real-Time and Embedded Systems

View full text Add to dashboard Cite

Large real-time software systems such as real-time Java virtual machines often use barrier protocols, which work for a dynamically varying number of threads without using centralized locking. Such barrier protocols, however, still suffer from priority inversion similar to centralized locking. We introduce gang priority management as a generic solution for avoiding unbounded priority inversion in barrier protocols. Our approach is either kernel-assisted (for efficiency) or library-based (for portability) but involves cooperation from the protocol designer (for generality). We implemented gang priority management in the Linux kernel and rewrote the garbage collection safe-point barrier protocol in IBM's WebSphere Real Time Java Virtual Machine to exploit it. We run experiments on an 8-way SMP machine in a multi-user and multi-process environment, and show that by avoiding unbounded priority inversion, the maximum latency to reach a barrier point is reduced by a factor of 5.3 and the application jitter is reduced by a factor of 1.5.

show abstract

Section: Jtres'09mentioning

confidence: 99%

Avoiding unbounded priority inversion in barrier protocols using gang priority management

Röck

Auerbach²,

Kirsch

et al. 2009

Proceedings of the 7th International Workshop on Java Technologies for Real-Time and Embedded Systems

View full text Add to dashboard Cite

show abstract

“…MACRO server-based agents employ novel services, such as the Spreading Activation Partial Order Planner (SA-POP) [11] and the Resource Allocation and Control Engine (RACE) [12]. These agents use the SA-POP service to (1) decompose goals into subgoals that are achieved at the server or by individual field nodes and (2) plan/schedule for their achievement.…”

Section: Overview Of Macromentioning

confidence: 99%

A flexible infrastructure for distributed deployment in adaptive sensor webs

Otte

Kinnebrew

Schmidt

et al. 2009

2009 IEEE Aerospace Conference

Self Cite

View full text Add to dashboard Cite

Abstract-Distributed sensor webs typically operate in dynamic environments where operating conditions, transient phenomena, availability of resources, and network connection quality change frequently and unpredictably. Often these changes can neither be completely anticipated nor accurately described during development or deployment. Our prior work has described how we developed agents and services that are capable of monitoring these changing conditions and adapting system parameters using the CORBA Component Model (CCM) deployment infrastructure as part of the Multiagent Architecture for Coordinated Responsive Observations (MACRO) platform.Our recent application of MACRO to the South East Alaska MOnitoring Network for Science, Telecommunications, Education, and Research (SEAMONSTER) project has identified new distributed deployment infrastructure challenges common to computationally constrained field environments in adaptive sensor webs. These challenges include standardized execution of low-level hardware-dependent actions and on-going data tasks, automated provisioning of agents for heterogeneous field hardware, and minimizing deployment infrastructure overhead. This paper describes how we extended MACRO to address these sensor web challenges by creating an action/effector framework standardizing the execution of lightweight actions and providing for automated provisioning of MACRO agents, in addition to footprint optimizations to the underlying CCM infrastructure.

show abstract

“…This case study is based on the Resource Allocation and Control Engine (RACE) [21], which is an open-source distributed resource manager system we developed using the CIAO [7] implementation of the Lightweight CORBA Component Model (CCM) [18] over the past several years in conjunction with Lockheed Martin and Raytheon. RACE deploys and manages Lightweight CCM application component assemblies (henceforth called operational strings) based on specifications of their resource availability/usage and QoS requirements [21]. Figure 2 shows the architecture of RACE, which is composed of four components assemblies (Input Adapter, Plan Analyzer, Planner Manager, and Output Adapter) that collaborate to manage operational strings for the target domain.…”

Section: The Race Dre System Case Studymentioning

confidence: 99%

CiCUTS: Combining System Execution Modeling Tools with Continuous Integration Environments

Hill

Schmidt

Porter

et al. 2008

15th Annual IEEE International Conference and Workshop on the Engineering of Computer Based Systems (Ecbs 2008)

Self Cite

View full text Add to dashboard Cite

System execution modeling (SEM) tools provide an effective means to evaluate the quality of service (QoS) of enterprise distributed real-time and embedded (DRE) systems. SEM tools facilitate testing and resolving performance issues throughout the entire development life-cycle, rather than waiting until final system integration. SEM tools have not historically focused on effective testing. New techniques are therefore needed to help bridge the gap between the early integration capabilities of SEM tools and testing so developers can focus on resolving strategic integration and performance issues, as opposed to wrestling with tedious and error-prone low-level testing concerns.This paper provides two contributions to research on using SEM tools to address enterprise DRE system integration challenges. First, we evaluate several approaches for combining continuous integration environments with SEM tools and describe CiCUTS, which combines the CUTS SEM tool with the CruiseControl.NET continuous integration environment. Second, we present a case study that shows how CiCUTS helps reduce the time and effort required to manage and execute integration tests that evaluate QoS metrics for a representative DRE system from the domain of shipboard computing. The results of our case study show that CiCUTS helps developers and testers ensure the performance of an example enterprise DRE system is within its QoS specifications throughout development, instead of waiting until system integration time to evaluate QoS.

show abstract

Design and Performance Evaluation of Configurable Component Middleware for End-to-End Adaptation of Distributed Real-Time Embedded Systems

Abstract: Standards-based quality of service (QoS)

Cited by 11 publications

References 22 publications

Avoiding unbounded priority inversion in barrier protocols using gang priority management

Avoiding unbounded priority inversion in barrier protocols using gang priority management

A flexible infrastructure for distributed deployment in adaptive sensor webs

CiCUTS: Combining System Execution Modeling Tools with Continuous Integration Environments

Contact Info

Product

Resources

About